It is well recognized that some drugs used in highly active antiretroviral therapy (HAART) of HIV infection have potentially adverse effects on human mitochondria, as exemplified by reports of the occurrence of life-threatening lactic acidosis. Although this entity is rare, the finding of mild lactic acidemia among patients on HAART suggests that lesser degrees of mitochondrial toxicity are more common than clinically recognized. Indeed, drug-induced defects in oxidative metabolism have been postulated to contribute to diverse toxicities of HAART. As the mechanisms and contribution of mitochondrial impairment to specific clinical toxicities become better defined, objective measures of oxidative dysfunction will be increasingly important. Thus, broadly applicable means of quantifying oxidative function in clinical populations are needed for use in clinical trials directed at reducing or reversing drug-induced mitochondrial toxicity. While morphologic and biochemical assessments of mitochondria involve invasive procedures and highly specialized techniques, rates of oxygen uptake (VO2) and carbon dioxide output (VCO2) determined from respired gases are non-invasive measures of oxidative metabolism. Exercise can be used as a provocative stress to increase the metabolic rate of skeletal muscle in a controlled manner, and the resulting changes in VO2 and VCO2 used to assess muscle oxidative capacity, as well as the relative contributions of oxidative and non-oxidative processes to the energy requirements of the imposed work. Exercise testing is therefore a logical approach to assessing mitochondria function, and systemic mitochondrial toxicity, on a systemic level. In addition, because exercise training itself represents a potential intervention for some toxicities associated with HAART, there is a pragmatic need to better understand both acute exercise responses and chronic adaptations to exercise training in this setting. The hypothesis of this proposal is that HAART-associated mitochondrial toxicity may be detected and quantified by measures of respiratory gas exchange reflecting oxidative metabolism during an exercise stress. To test this there are two specific aims: 1) To demonstrate the ability of exercise testing to identify abnormalities in oxidative function in a cross-section of HIV infected patients, and identify test variables that best discriminate between patients with and without HAARTassociated mitochondrial dysfunction. 2) To validate the utility of exercise testing for monitoring changes in oxidative function in patients receiving HAART and to reflect important cellular responses to treatment interventions, using longitudinal assessments of patients before and after an exercise training intervention.